The project Evion had brought to Twitter the possibility of real time matching of tweets with relevant respondents, aiming to create real time feeds of high relevancy Twitter updates personalized using users' individual interests. Evion used machine learning and crowd filtering algorithms to find in real time the new information on Twitter that could be interesting for its users, and relied on Twitter's @mentions for delivery.
Evion ran on Twitter during the second half of 2011, quickly gaining over 6500 followers. Due to technical disruptions from Twitter's side, however, the service was stopped in January of 2012. For more information, you can examine the past Evion's activity from the links below.
Evion Twitter Accounts:
Evion had been developed in 2010-2011 by Yuriy Mishchenko, at the time a postdoctoral fellow at the Department of Statistics at Columbia University, and Ushananthan Ganeshananthan, at the time an MBA student at Columbia Business School.